Control and anti-control of chaos based on the moving largest Lyapunov exponent using reinforcement learning

In this work, we propose a method of control and anti-control chaos based on the moving largest Lyapunov exponent using reinforcement learning. method, design reward function for learning according to exponent, which is similar average but computes corresponding recently updated time series with fixed, short length. We adopt density peaks-based clustering algorithm determine linear region divergence index so that can obtain small data set by fitting slope region. show proposed fast easy implement through controlling anti-controlling typical systems such as Henon map Lorenz system.